MI355X FP4: M3 EAGLE vs Off Speculative Decoding
Speculative decoding comparison of M3 EAGLE versus Off on MI355X FP4 (AMD CDNA 4) running MiniMax M3 428B. Throughput, cost, and interactivity differences across LLM workloads. Use the chart controls below to switch sequences and metrics — same interactions as the main inference chart.
MTP acceptance-rate implementations differ across inference engines. Points from different engines are not directly comparable on the same curve — throughput and cost at matched interactivity may reflect engine-level differences rather than pure speculative decoding gains. Interpret cross-engine comparisons with caution.
M3 EAGLE posts 3263 tok/s/GPU for $0.13 per million tokens at 65 tok/s/user on MiniMax M3 428B (MI355X FP4); Off posts 1738 tok/s/GPU for $0.24. M3 EAGLE is 90% cheaper per token; M3 EAGLE delivers 88% more tok/s/GPU. Draft-token acceptance rates determine whether speculative decoding helps or hurts at a given concurrency level.
Throughput at 104 tok/s/user on MiniMax M3 428B (MI355X FP4): M3 EAGLE hits 2260 tok/s/GPU, Off hits 756. Per-million costs land at $0.18 and $0.55 respectively. M3 EAGLE is 200% cheaper per token; M3 EAGLE delivers 199% more tok/s/GPU. Speculative decoding trades extra compute on draft tokens for fewer decoding steps — the payoff depends on sequence length and batch size.
Toward the upper edge of the 27–181 tok/s/user interactivity band, at 143 tok/s/user on MiniMax M3 428B (MI355X FP4): M3 EAGLE runs 1577 tok/s/GPU at $0.26/M tokens, Off runs 242 at $1.70/M. M3 EAGLE is 547% cheaper per token; M3 EAGLE delivers 551% more tok/s/GPU. Gains from speculative decoding vary by workload; short-output prompts tend to benefit less. (Numbers reflect this URL's pinned 1k/1k · fp4 workload — changing sequence or model updates both the table and chart; the table stays pinned to this page's precision, so precision toggles in the controls affect the chart only.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/gpu) | M3 EAGLE:3263.3Off:1737.6 | M3 EAGLE:2259.9Off:755.7 | M3 EAGLE:1577.3Off:242.4 |
| Cost ($/M tok) | M3 EAGLE:$0.126Off:$0.239 | M3 EAGLE:$0.181Off:$0.545 | M3 EAGLE:$0.262Off:$1.698 |
| tok/s/MW | M3 EAGLE:1231437Off:655695 | M3 EAGLE:852779Off:285158 | M3 EAGLE:595217Off:91458 |
| Concurrency | M3 EAGLE:~105Off:~55 | M3 EAGLE:~46Off:~15 | M3 EAGLE:~24Off:~4 |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.